Editor’s Note:

Hepatitis B virus (HBV) infection is a widespread global health issue. According to the World Health Organization, as of 2019, approximately 296 million people globally suffer from chronic HBV infection, with around 1.5 million new infections annually, leading to 820,000 deaths due to HBV-related complications. Data from the Chinese Center for Disease Control and Prevention indicates that around 86 million people in China have chronic HBV infection. Simultaneously, with changes in lifestyle and dietary habits, metabolic-associated fatty liver disease (MAFLD) has become the most common chronic liver disease globally, affecting approximately one-third of the world’s population. In Asia, the prevalence of MAFLD has been steadily rising, increasing from 25.28% between 1999 and 2005 to 33.90% between 2012 and 2017.

Chronic HBV infection combined with MAFLD is increasingly common in clinical practice. However, the risk factors affecting the prognosis of patients with HBV infection and concurrent metabolic-associated fatty liver disease are not yet clear. New diagnostic models to assist in clinical diagnosis and grading still need to be established. At the 74th Annual Meeting of the American Association for the Study of Liver Diseases (AASLD 2023), held in Boston from November 10 to 14, the team of Professors Chao Wu and Jie Li from the Department of Infectious Diseases at Nanjing University Medical School’s Drum Tower Hospital presented several clinical research abstracts on non-invasive diagnosis, risk factors, and treatment management of liver diseases, achieving fruitful outcomes. “Hepatology Digest” hereby reports on the relevant content as follows.

▲ Professor Jie Li’s team at AASLD 2023

01 Non-invasive Diagnosis – Fibroscan

Study 1: Determination of the Optimal Transient Elastography Threshold for Progression of Liver Fibrosis in Patients with Chronic Hepatitis B Complicated by Hepatic Steatosis

Key Findings: Involving 869 patients with chronic hepatitis B complicated by hepatic steatosis who underwent liver biopsy and Fibroscan, 74 patients were found to have progressing fibrosis. The recommended transient elastography (TE) threshold for chronic hepatitis B or NAFLD patients was 9.7 kPa, with an corresponding area under the receiver operating characteristic curve (AUROC) of 0.681, sensitivity (SE) of 55.41%, specificity (SP) of 80.71%, positive predictive value (PPV) of 28.30%, and negative predictive value (NPV) of 92.90%. At the optimal threshold of 8.8 kPa, the corresponding AUROC was 0.733, significantly higher than 0.681 (P=0.015). The SE, SP, PPV, and NPV for the optimal threshold (8.8 kPa) were 70.27%, 76.25%, 28.9%, and 94.9%, respectively, surpassing the existing threshold (9.7 kPa). Subgroup analysis showed that for patients with BMI < 25 kg/m2, the AUROC for the optimal threshold (8.8 kPa) was 0.708, significantly higher than the existing threshold (AUROC=0.566) (P=0.003). For patients with BMI ≥ 25 kg/m2, the AUROC for the optimal threshold was 0.742, exceeding the existing threshold (AUROC=0.737), though not significantly different (P=0.747).

Main Conclusion: The optimal TE threshold for assessing the progression of fibrosis in patients with chronic hepatitis B complicated by hepatic steatosis is 8.8 kPa, lower than the 9.7 kPa used for chronic hepatitis B alone. In patients with chronic hepatitis B and hepatic steatosis, a stricter TE value is needed for a more accurate assessment of fibrosis progression.

Abstract: Rui FJ, Xu XM, Ni WJ, Xu L, Liang J, Yang YH, Li J. Determining the optimal transient elastography cutoffs in chronic hepatitis B patients with concurrent hepatic steatosis. AASLD 2023, Abstract (3455-A).

Study Two: Optimal LSM Diagnostic Threshold for Significant Liver Fibrosis in Patients with Chronic Hepatitis B

Key Findings: A total of 332 patients with chronic hepatitis B who underwent liver biopsy and liver transient elastography were included. The training set comprised 145 patients, of which 79 had significant fibrosis. Internal and external validation sets included 31 and 38 patients with significant fibrosis, respectively. There was a strong correlation between LSM and the severity of fibrosis. In the training set, the predicted AUC for LSM for diagnosing significant fibrosis was 0.753, 0.839 in the internal validation set, and 0.811 in the external validation set. The determined optimal LSM threshold for diagnosing significant fibrosis was 9.8 kPa, with a specificity of 95.5%, PPV of 93.3%, and a misdiagnosis rate of 7.6%, significantly better than the 2020 East Asian expert-recommended threshold of LSM ≥ 8 kPa, with a specificity of 74.2%, PPV of 73.8%, and a misdiagnosis rate of 25.8%. Additionally, in the internal validation set, the optimal threshold’s specificity was 96.8%, and PPV was 93.8%, while in the external validation set, specificity was 92.1%, and PPV was 94.0%. The misdiagnosis rates for the newly determined LSM threshold were all below 10% in all three cohorts. However, finding optimal LSM thresholds to exclude significant fibrosis or diagnose/exclude cirrhosis was challenging.

Main Conclusion: The newly determined and validated LSM threshold of ≥9.8 kPa for diagnosing significant liver fibrosis has a low misdiagnosis rate and contributes to the diagnosis and treatment of chronic hepatitis B patients.

Abstract: Zhang ZY, Zhu L, Liu XX, Wang J, Zhang SQ, Yan XM, Qiu YW, Li J, Huang R, Wu C. Optimized cutoff of liver stiffness measurement for ruling in significant fibrosis in patients with chronic hepatitis B. AASLD 2023, Abstract (1324-C).

02 Non-invasive Diagnosis – Combined Model

Study Three: Validation of the PAP Model’s Diagnostic Efficiency for Progressive Liver Fibrosis and Cirrhosis in Different Subgroups of Chronic Hepatitis B Patients with Concurrent Hepatic Steatosis

Key Findings: The team previously developed a PAP diagnostic model to assess progressive liver fibrosis and cirrhosis in chronic hepatitis B patients with concurrent hepatic steatosis. PAP, based on Gaussian Naive Bayes, incorporates prothrombin time, albumin, and platelet count for model construction. A total of 1427 patients from nine clinical centers in China were retrospectively included, and patients were divided into five groups in the training and validation sets based on age (≥40 years and <40 years), gender (male and female), alanine aminotransferase (ALT) levels (normal and elevated), hepatitis B e antigen (HBeAg) status (positive and negative), and HBV DNA levels (<105 and ≥105 IU/mL).

In the training set (n=1063), the PAP model achieved the highest area under the curve (AUC) for diagnosing progressive liver fibrosis (≥S3) in patients with HBV DNA ≥105 IU/mL, reaching 0.827 (95% CI: 0.749–0.905). For diagnosing cirrhosis (≥S4), the PAP model performed best in patients aged ≥40 years, with an AUC of 0.864 (95% CI: 0.794–0.934). However, its performance in diagnosing progressive liver fibrosis and cirrhosis in patients with normal ALT levels was suboptimal.

In the validation set (n=364), the PAP model achieved the highest AUC for diagnosing progressive liver fibrosis (≥S3) in patients with HBV DNA ≥105 IU/mL, at 0.824 (95% CI: 0.765–0.883). The AUC was lowest in patients aged <40 years, at only 0.687 (95% CI: 0.619–0.755). For diagnosing cirrhosis (≥S4), the PAP model performed best in female patients, with an AUC of 0.954 (95% CI: 0.902–1.000), while its AUC was lowest in patients with HBV DNA <105 IU/mL, at 0.771 (95% CI: 0.708–0.834). http://py.reallife-liver.com/ is a free online calculator for PAP model fibrosis scoring.

Main Conclusion: In patients with chronic hepatitis B and concurrent hepatic steatosis, the PAP model demonstrates repeatability in diagnosing progressive liver fibrosis and cirrhosis in different populations, offering potential as a new diagnostic tool.

Abstract: Ni WJ, Yeo YH, Rui FJ, Xu YY, Xu L, Zheng Q, Tian XR, Zeng QL, He ZB, Qiu YW, Zhu CW, Ding WM, Wang J, Huang R, Xue Q, Wang XQ, Qi XL, Shi JP, Wu C, Liang J, Li J. Validation of the PAP model performance for diagnosing advanced fibrosis and cirrhosis in chronic hepatitis B patients with concurrent hepatic steatosis across different subgroups. AASLD 2023, Abstract (1332-C).

Study Four: Simplifying the Liver Fibrosis Diagnostic Process to Reduce Uncertain Cases: A New Diagnostic Model for Patients with Chronic Hepatitis B and Concurrent Non-alcoholic Fatty Liver Disease

Key Findings: A total of 926 untreated patients with chronic hepatitis B and concurrent NAFLD from nine Chinese medical institutions were included, with 180 patients (19.44%) having advanced fibrosis. At lower cutoff values, APRI had the highest sensitivity at 65.56%, followed by FIB-4 (53.89%) and NFS (41.11%). At higher cutoff values, NFS had the highest specificity at 97.99%, followed by FIB-4 (94.77%) and APRI (94.1%).

Notably, compared to other scores, the ratio of uncertain cases diagnosed by APRI for progressive liver fibrosis was the highest at 30.99%. When using a single non-invasive score, correct classification rates for NFS, FIB-4, and APRI were 64.7%, 64.1%, and 57.6%, respectively. When using a combined non-invasive score, the NFS-APRI combined model achieved a correct classification rate for 73.7% of patients, better than FIB-4-NFS (71.3%), NFS-FIB-4 (71.8%), FIB-4-APRI (71.0%), APRI-FIB-4 (70.5%), and APRI-NFS (71.8%).

Main Conclusion: The NFS-APRI combination model demonstrates superior performance in improving the discrimination of progressive liver fibrosis, reducing uncertain cases, and optimizing referral pathways, providing a non-invasive diagnostic basis for targeted liver fibrosis screening.

Abstract: Xu XM, Rui FJ, Ni WJ, Wu C, Liang J, Shi JP, Yeo YH, Li J. Streamlining the diagnosis of liver fibrosis to reduce patients with indeterminate results: A novel algorithm for CHB patients with concurrent NAFLD. AASLD 2023, Abstract (1328-C).

03 Risk Factors

Study Five: Diabetes Mellitus as an Independent Predictive Factor for Significant Inflammation and Fibrosis in Chronic Hepatitis B Patients with Concurrent Hepatic Steatosis

Key Findings: A total of 869 patients with chronic hepatitis B and concurrent hepatic steatosis who underwent liver biopsy at eight medical centers in China were included. Among them, 8.2% (71 patients) had diabetes mellitus (DM), with an average body mass index (BMI) of 24.9±3.3 kg/m2 and an average HBV DNA of 5.3±2.0 log10 IU/mL. Approximately half of the patients (380 patients, 46.3%) were HBeAg positive, and 5.9% of patients (42 patients) were undergoing antiviral therapy. Moderate to severe hepatic steatosis (grades 2-3) was present in 24.3% of patients (206 patients). Most patients (529, 60.9%) had significant inflammation (G2-4), and about half of the patients (431, 49.6%) had significant fibrosis (F2-4). Compared to non-diabetic patients, diabetic patients were more likely to have significant inflammation (76.1% vs. 59.7%, P=0.02) or significant fibrosis (76.1% vs. 47.3%, P<0.001).

In multivariate logistic analysis adjusting for age, gender, BMI, HBV DNA, and HBeAg (Model 1), DM was independently associated with significant inflammation (OR: 3.60; 95% CI: 1.56–8.33; P=0.003) and significant fibrosis (OR: 4.08; 95% CI: 1.95–8.53; P<0.001). After further adjusting for hepatic steatosis (Model 2), DM remained independently associated with significant inflammation (OR: 3.38; 95% CI: 1.46–7.86; P=0.005) and significant fibrosis (OR: 4.49; 95% CI: 2.08–9.72; P<0.001).

Main Conclusion: CHB patients with concurrent hepatic steatosis and DM are more than three times likely to experience significant liver inflammation and more than four times likely to develop significant fibrosis compared to non-diabetic patients. Since liver inflammation and fibrosis are major predictors of future liver-related outcomes, collaboration between liver and metabolic specialists is needed to optimize the management of chronic hepatitis B patients with hepatic steatosis and diabetes.

Abstract: Li J, Rui FJ, Nguyen B, Zheng Q, Zeng QL, He ZB, Shi JP, Wu C, Nguyen MH. Diabetes mellitus is an independent predictor of significant inflammation and fibrosis in chronic hepatitis b patients concurrent with hepatic steatosis. AASLD2023, Abstract (1309-C)

04 Treatment Management

Study Six: Implementing the Comprehensive Intervention Strategy (GATE) Plan to Increase Antiviral Treatment Rates in Patients with Chronic Hepatitis B

Key Findings: The study established a retrospective cohort and a prospective cohort, including chronic hepatitis B patients who did not receive antiviral treatment between January 2019 and October 2021 and between November 2021 and February 2023, respectively. Before implementing the Comprehensive Intervention Strategy (GATE), the retrospective cohort included 1134 CHB patients meeting treatment criteria, with 71.08% of patients receiving antiviral treatment. Among patients meeting treatment criteria, 22.05% were not advised by doctors to undergo antiviral treatment, and 6.88% refused treatment due to various personal concerns, including concerns about treatment duration, efficacy, adverse reactions, high costs, and family planning. After the implementation of the comprehensive intervention strategy, the prospective cohort included 252 patients, and the antiviral treatment rate increased by 8.68% (from 71.08% to 79.76%, P=0.005). The proportion of patients not currently recommended antiviral treatment decreased from 22.05% to 10.71% (P<0.001).

Main Conclusion: The establishment of a comprehensive intervention strategy, including online and offline education and the design of a follow-up reminder system, can narrow the gap between CHB patients eligible for treatment according to guidelines and those actually treated, thereby increasing the treatment rate.

Abstract: Zhang SQ, Wang J, Yan XM, Wang L, Zhang ZP, Lu SF, Chen YX, Li J, Huang R, Wu C. Interim results of a program to narrow gap between in accordance with guidelines and consent to treat CHB population in east of China (GATE STUDY). AASLD2023, Abstract (1364-C).

The poster presentations of Professor Chao Wu and Professor Jie Li’s team at AASLD 2023